Moving from Item Rating to Features Relevance in Top-N Recommendation
Authors
Anelli Vito Walter, Di Noia Tommaso, Di Sciascio Eugenio, Lops Pasquale, Trotta JosephAbstract
Although very effective in computing accurate recommendations, due to their inner nature, collaborative algorithms work very well with dense matrices but show their limits when they deal with sparse ones. In these cases, using only past ratings may lead to unsatisfactory results in the recommendation list. In this paper we show how to move from a user-item to a user-feature matrix by exploiting original user ratings. We then use matrix factorization techniques to compute recommendations.
DOI
https://doi.org/10.1007/978-3-030-15712-8_63BibTex references
@InProceedings{ADDLT18, author = "Anelli, Vito Walter and Di Noia, Tommaso and Di Sciascio, Eugenio and Lops, Pasquale and Trotta, Joseph", title = "Moving from Item Rating to Features Relevance in Top-N Recommendation ", booktitle = "Proceedings of the 9th Italian Information Retrieval Workshop - IIR", volume = "2140", year = "2018", publisher = "CEUR-WS", url = "http://sisinflab.poliba.it/Publications/2018/ADDLT18" }